The class of "rumpo" verbs plus directional prefixes

(1)

Lingue antiche e moderne 9 (2020) ISSN 2281-4841 DOI: 10.4424/lam92020-10

The class of rumpo verbs plus directional prefixes

Cadina Palachi

T

HE CLASS OF RUMPO VERBS PLUS DIRECTIONAL PREFIXES

C

^{A INA}

P

^ALACHI

ABSTRACT

The Latin verb rumpo can be classified, according to Levin (1993), as a break verb. This class belongs to a larger class which e presses change of state . The Latin verb frango appears to have a similar meaning to that of rumpo. Both verbs are transitive and can form new ones with prepositional prefi es. However, in the presence of prefi es that indicate direction (such as in, ex or intro), the result of the derivation is different for each one. Frango retains its meaning and the direct complement is the entity that changes state , while rumpo modifies its meaning, turns into a movement verb and becomes intransitive, its main component being direction. The uestion we will attempt to answer is: why is the meaning of direction – present in the prefi – favoured in the new le emes from rumpo, while in the prefi ed verbs from frango is the feature cause – apparently present in the change of state root – preserved The hypothesis is that the roots RUMP and FRANG have different meanings; the meaning of movement is present in the root RUMP while FRANG is closer to the verbs belonging to the destroy class. Our analysis describes the semantics and synta of these verbs in the theoretical framework of istributed Morphology and focalizes in the problem related to the content of Roots.

(2)

Cadina Palachi 226

Lingue antiche e moderne 9 (2020) ISSN 2281-4841

1. T

HE MEANING OF ERBAL

R

OOTS

AN ITS CONTRIBUTION TO WOR FORMATION

1.1. The problem

This paper will try to account for the differences in meaning and synta between the Roots RUMP and FRANG FRING in the simple verbs rumpo and frango, and in the derivative verbs with prefi es in and ex (irrumpo, erumpo, infringo, effringo)

¹

. When prefi ed with in and ex , frango retains its meaning and the direct complement – the entity that changes state – while rumpo modifies its meaning, turns into a movement verb and becomes intransitive, its main component being direction. There is another problem to which we would like to draw attention: the literature on this subject ( lvarez Huerta 2009) mentions that the prefi transitivizes the verb, whereas this is the opposite case, a transitive verb (rumpo) becomes intransitive (irrumpo, erumpo).

The Latin verb rumpo, according to the Oxford Latin Dictionary (OLD), is transitive and it means: to cause to split open or e plode, to cause to break, to burst (barriers, bonds) as to escape . Frango has a similar meaning, according to the OLD, To break, shutter, smash, to break in pieces, split up, to break open, break down (barriers), to break through, penetrate .

These verbs, according to the classes proposed by Levin (1993), may be considered break verbs and they enter into a larger class that e presses change of state . Levin (1993: 242) says (following Hale – eyser 1987) that break verbs refer to actions that bring about a change in the ‘material integrity’ of some entity . Rumpo and frango, as defined by the OLD, are transitive verbs that e press the destruction

1 Another problem, that we are not going to address here, is the identification of the allomorphs of a single Root. Harley mentions that Factors that a linguist might take into account in identifying occurrences of a root across different conte ts include identity or similarity of form, identity or similarity of meaning, and purely morphological behaviors, such as idiosyncratic selectional restrictions with respect to affi ation or other morphological processes (Harley 2014: 1).

(3)

The class of rumpo verbs plus directional prefixes 227

or degradation of an entity, material or immaterial, designated by the direct complement ( P in accusative). This P-Acc changes its state (it’s aggravated), or ceases to e ist as such (as can be seen in 1a and b).

These verbs contain a cause feature and, when the e ternal argument is an agent, they admit the presence of an instrument in ablative (sacrifica securi, in 1a).

(1) veluti qui candida tauri

rumpere sacrifica molitur colla securi. (Ov. met. 12, 249) As when one strives to break a bull’s white neck with sacrificial a e . (Trans. F.J. Miller)

(2) mollis illa educatio, quam indulgentiam uocamus, neruos omnis mentis et corporis frangit. ( uint. inst. 1, 2, 6, 4)

That soft upbringing, which we call kindness, saps all the sinews both of mind and body . (Trans. E. Butler)

The classification of these verbs into the break class could work well for both simple verbs, but it is problematic when they form new derivative verbs with prepositional prefi es. The meaning of the derivatives, with the prefi es that indicate direction in and ex , does not remain constant for rumpo. This verb no longer indicates a change of state in an entity (there is no entity that is destroyed or that suffers some type of change in its material integrity ). Rumpo with prefi es (irrumpo, erumpo) e presses movement and change of location of the e ternal argument. It becomes intransitive and has no direct complement in accusative. A P-Acc could be present, but, when it is, it e presses not the object that changes its state, but the Ground of Movement

²

. Moreover, we can find a directional PP in Acc. or ex Abl. (as it can be seen in 3a, inrumpat in pectora). The main component

2 We use ground in the sense of Talmy (2000: 312): The Ground is a reference entity, one that has stationary setting relative to a reference frame, with respect to which the Figure’s path, site or orientation is characterized .

(4)

Cadina Palachi 228

of meaning of the derived verbs (irrumpo, erumpo) is IRECTION, specifically, the boundary crossing (outwards or inwards). The verb frango with prefi es retains its meaning of breaking or destroying . The verbs derived from frango (infringo, effringo) re uire a direct complement, a P-Acc., that e presses the entity that changes state or that is destroyed (as it can be seen in 3b, infregit).

(3) a. ut denique dominetur in adfectibus atque in pectora inrumpat. ( uint. inst. 2, 5, 8, 6)

how the orator stablishes his sway over the emotions of the audience, forces his way into their hearts . (Trans. E.

Butler)

b. rigidum fera dextera cornu dum tenet, infregit, truncaque a fronte revellit.

(Ov. met. 9, 86) holding my tough horn in his pitiless right hand, he broke it off and tore it from my forehead, mutilating me . (Trans. F.J. Miller)

The uestion we will try to account for is what kind of meaning the Root has and how does it change in conte t or according to the type of relational elements in which they are inserted (as used in Acedo- Matellán – Mateu 2014).

This is to say why is the meaning of IRECTION – associated with the prefi – favoured in the derivations of prefi ed rumpo, while in the prefi ed frango the meaning of CHANGE OF STATE, as in the simple verb, is preserved

We believe it is necessary to observe how the meaning of the prefi ed

verb is built from the union of the relational elements in which the Roots

are inserted and the meaning of the Roots. This brings us to the uestion

about the meaning that is associated with the Roots and also to consider

the way they are distributed in the synta . Why can RUMP be inserted

in structures of CHANGE OF LOCATION and FRANG only in those

(5)

of CHANGE OF STATE The hypothesis is that rumpo and frango are not synonymous and that the meaning of frango is nearer to the class of destroy.

The present paper is organized as follows: Section 1 sets out the research uestion and some specifications of the theoretical framework;

Section 2 describes the meaning of the simple verbs rumpo and frango;

Section 3 e poses the changes that occur when the Roots of these verbs enter the derivation with in and ex ; Section 4 describes the type of Ps accusatives that can be complements of the prefi ed verbs (infringo, effringo) and Section 5 presents the conclusions and new research uestions.

The e amples were taken from a corpus that corresponds roughly to the te ts of classical Latin. We have selected some authors that are considered representative of a Latin stage, with the e ception of Plautus, which was included because of our interest in comparing certain structures of him with others that belong to the classical corpus:

Cicero, Caesar, uintilian, irgil, Livy, Catullus, Lucretius, Ovid.

1.2. Specifications about the framework

We assume that word formation takes place in synta and that the primitives combined are: Roots and terminal nodes (l-morphemes and f-morphemes, Harley – Noyer 1999). A central debate of morphological theory consists in defining what kind of meaning the Roots add and, in close relation to this problem, whether or not they can take arguments.

Arad (2005: 1) defines the Roots as atomic le ical elements . Within the framework of istributed Morphology (Halle – Marantz 1993;

Marantz 1997; Harley – Noyer 1997; Harley 2014; Embick 2015),

meaning can be understood as disjoint or bifurcated and combining

the elements provides different components of meaning. In other words,

the meaning of the Roots is built conte tually: according to Embick

(2009: 1) there are at least four components to meaning: the meaning

of Roots, the denotations of functional heads in a syntactic structure,

(6)

Cadina Palachi 23 0

operators introduced in the semantics (CAUSE, BECOME, etc.) when syntactic structures are interpreted, and aspects of meaning that arise via (competition for) use

³

.

Within this framework, it is necessary to ask ourselves what kind of meaning the Roots can carry. Embick (2009: 1) also asks this uestion:

What’s in a Root (M1) I.e, what types of semantic information could be built into the semantics of a Root

⁴

.

Acedo-Matellán – Mateu (2014) argue that there are two types of elements that merge in the synta to form the argument structure (which, therefore, is not primitive): relational and non-relational elements. The non-relational elements are Roots and argumental Ps.

Only relational elements can project structure (Acedo-Matellán – Mateu 2014: 15). In this theory, Roots have semantic properties that are not relevant to the synta , while the relational structures have semantic features that are relevant.

In the definition of the Motion event, we follow Talmy (2000: 25) the basic Motion event consists of one object (the Figure) moving or located with respect to another object (the reference object or Ground) . The other components of the Motion event are: Path (the path where the Figure moves or is located), Manner and Cause.

3 According to the Bifurcation Thesis for Roots (BT-R): If a component of meaning is introduced by a semantic rule that applies to elements in combination, then that component of meaning cannot be part of the meaning of a Root (Embick 2009: 1).

4 Another problem that arises in relation to the Roots is whether they have a phonological representation from the beginning of the derivation (Embick 2005) or if they are subject to late insertion and therefore are ocabulary Items. It is also possible to claim, as Borer (2013) does in fact, that the Roots have just phonological representation during the derivation. In this work we will not deal with this problem, although it is true that FRANG presents, in the verbs that interest us, an allomorph whose phonological representation is f ; in that sense, It can be thought that is necessary to postulate the late insertion of the root in order to solve the problem of allomorphy. But we will not address this argument because we will focus on problems of meaning.

(7)

The class of rumpo verbs plus directional prefixes 23 1

2. T

HE SIMPLE ERBS FRANGO AN RUMPO

2.1. Description

In this section we will describe the syntactic structures of the non- prefi ed verbs in order to show that frango has a similar meaning to rumpo, but cannot be strictly classified within the break class. We will also outline the type of arguments that are possible to occur with the simple verbs formed from these Roots.

Beth Levin’s (1993) classification of the English verbs distinguishes break verbs, which are included in the change of state verbs, from destroy verbs. In Latin there is a set of destroy verbs that relate to the total destruction of the entities (Levin 1993: 239): destruo destroy

⁵

, pereo pass away, disappear , diruo demolish, destroy , subruo undermine , demolior remove, destroy . The break verbs class, instead, describes specifics of the resulting physical state of an entity (e.g. whether something is broken, splintered, cracked, and so on) rather than simply describing the fact that it is totally destroyed (Levin 1993:

239). In Latin, the set of break verbs includes verbs like scindo cut, split , solvo loose, dissolve , divido separate and rumpo. The uestion is what differentiates rumpo from frango. Is it better to classify frango as a destroy verb

In (4) we present some e amples of frango, and in (5) of rumpo. The state of broken can be achieved by the action of a P originator in nominative case: Martia legio (in 4a), Octavius (in 4b), the greatness of Cesar’s spirit and the office of the Senate (in 4c), and the one who wants to eat the kernel (in 4d); so that the causes are not only human, or physical, but also abstract Ps. It should be noted that in (4b) Octavius is the originator of the action, but the instrument is his patience ( P- Abl.).

Another important uestion is what kind of entities can be turned into a broken state. It should also be mentioned that it is possible for these

5 efinitions in brackets were taken from the Oxford Latin Dictionary (OLD).

(8)

Cadina Palachi 23 2

entities to be broken to a lesser or bigger e tent . In the e amples of (4) the Ps in accusative are: furorem (4a), Tiberium Gracchum (4b), meum consilium (4c) and nucem (4d).

(4) a. post eiusdem furorem Martia legio fregit, quarta adflixit.

(Cic. Phil. 10, 21, 4)

Afterwards the Martian legion has broken his frenzy and the forth crashed it to the earth . (Trans. W. er)

b. is, qui iniuria accepta fregit Ti. Gracchum patientia, civis in rebus optimis constantissimus M. Octavius. (Cic. Brut.

95, 6)

M. Octavius, a man of infle ible constancy in every just and laudable measure; and who, after being affronted and disgraced in the most public manner, defeated his rival Tiberius Gracchus by the mere dint of his perseverance . (Trans. E. Jones)

c. fregit hoc meum consilium et Caesaris magnitudo animi et senatus officium. (Cic. Fam. 4, 4, 5)

This determination of mine suddenly gave away before Caesar’s magnanimity and the Senate’s devotion . (Trans. G. Williams)

d. qui e nuce nuculeum esse volt, frangit nucem. (Plaut.

Curc. 55)

The man that wants to eat the kernel, craks the shell . (Trans. P. Ni on)

In the e amples of (5), structures where the root RUMP is inserted, the elements that originate the change of state are Ps in nominative:

nulla dies (in 5a); ingens pavor (in 5b); caelatura (the chisel hit) (in

5c); qui (the gods) (in 5d). And it is also possible to observe the type of

Ps in accusative that end up broken to a lesser or bigger e tent :

(9)

peace and pacts (in 5a); sleep (in 5b); the sheet too thin (in 5c.) and the decrees of the ancient sisters (in 5d); which are all of very different kind (concrete and abstract things).

(5) a. nulla dies pacem hanc Italis nec foedera rumpet. ( erg.

Aen. 12, 202)

No dark day shall cause the Italians to shatter thes treaty’s peace-terms . (Trans. F. Ahl)

b. olli somnum ingens rumpit pauor. ( erg. Aen. 7, 458) Frightful and massive shock bursts in on his sleep . (Trans. F. Ahl)

c. erit autem, si non ab initio tenuem nimium laminam duxerimus et quam caelatura altior rumpat. ( uint. inst.

2, 4, 8, 1)

And there will be sufficient if only we do not draw the plate too thin to begin with, so that it run the risk of being broken if the graver cut too deep . (Trans. Buttler)

d. Talia nequiquam toto Venus anxia caelo verba iacit superosque movet, qui rumpere quamquam ferrea non possunt veterum decreta sororum. (Ov. met. 15, 780)

The an ious goddess cried these complaints throughout the sky, but all in vain. The gods were moved indeed; and although they were not able to break the iron decrees of the ancient sisters . (Trans. F.J. Miller)

In conclusion, it could be said that the simple verbs formed from the Roots RUMP and FRANG are almost synonymous, with an argument in nominative as the originator of the event and a P-Acc.

that is broken or, at least, degraded in its characteristics. There is a

possibility that a P-Abl. e presses the instrument.

(10)

Cadina Palachi 23 4

Although both verbs seem to be synonymous, we will see that they are not. There are some differences between the resulting states of the objects e pressed by the Ps-Acc., that end up broken in the structures of rumpo; and the Ps-Acc. in Ps with frango. In (6), it is possible to observe that with rumpo there are Ps-Abl. (or, as in 6b, with the preposition ab a P-Abl.) that indicate the Paths, as in (6a) multis locis; or the part of the object that is broken: in (6b), a pectore. In (6c), the rupta via is the Path that allows us (nos venerimus) to arrive to the clausula.

(6) a. rupti inde multis locis ordines. (Liv. 6, 13, 3)

Then, the ranks were broken in many places . (Trans.

Radice)

b. tenues a pectore vestes rumpit. (Ov. ars 3, 708)

she tore the thin clothing from her breast . (Trans. line) c. gratiam perdet si ad eam (clausulam) rupta via

venerimus. ( uint. inst. 9, 4, 63, 4)

its elegance will be wasted, if the path which leads up to it be interrupted . (Trans. Butler)

In these e amples it can be seen that the resulting broken state with rumpo does not consist in the total destruction of the entity. The object results broken in some parts, and those parts become the Path of Motion for the Figure.

Although we cannot say that this behavior is what Levin (1993) calls Conative alternation , we think there are some similarities. The instrument makes contact with the object and moving makes it permeable

⁶

. If we look back on the e amples of frango in (4), we can

6 Levin (1993) distinguishes cut verbs from break verbs. One of the differences between the two classes is the possibility of cut verbs to enter the Conative alternation. This alternation is defined by Levin (1993: 42) as a transitive

(11)

see that the objects end up totally broken or destroyed. There is no creation of Paths or places where the Figure could move through.

2.1. Analysis

The resulting verbs, beyond being apparently simple from a morphological point of view, have several layers of functional meanings, at least one operator, a vP layer, which e presses the CAUSE or the ORIGIN; and one of CHANGE, a pP layer. Roots are inserted into such relational structures and provide the form of the resulting state.

For our analysis we follow Acedo-Matellán – Mateu (2014) who proposed that the meaning of the Roots and the arguments depend on the place they occupy in the proposed relational structure. Thus, the vP specifier is an originator, the pP specifier is the Figure and the double pP complement is the Final Result or Final Ground. For these verbs, the structure we propose, following Acedo-Matellán – Mateu (2014), is:

alternation in which the object of the verb in the transitive variant turns up in the intransitive conative variant as the object of the preposition in a prepositional phrase headed by the preposition at The conative alternation seems to be found with verbs whose meaning includes notions of both contact and motion .

(12)

Cadina Palachi 23 6

vP P v’

ui Octavius

v pP

P p’

ferrea decreta T. Gracchum

p pP

p RUMP FRANG Qui rumpere quamquam ferrea non possum veterum decreta sororum. (Ov. met. 10, 36)

Fregit Ti. Gracchum Octavius (Cic. Brut. 95)

3.

THE PREFI E ERBS INFRINGO EFFRINGO S

.

IRRUMPO ERUMPO

3.1. Description

The Roots FRANG and RUMP also have the possibility of being inserted into structures with prefi es in or ex . However, the meaning of the Roots in infringo effringo is different than the ones in irrumpo/erumpo and the same happens with the types of arguments re uired in each case.

erbs formed from frango enter syntactic structures with Ps in

nominative that are the originators of the change of state. In (7) we

provide e amples for infringo (break, crush in or on): hoc proelium

(7a); in (7b), the originator is not e pressed, but it can be inferred from

(13)

the conte t that is Attalus’ luck (fortuna Attali); in (7c), Hercules

⁷

, in which a P-Abl. e presses the instrument (fera dextera). The Ps-Acc.

represent the objects that change of state: (7a) res Samnitium; in (7b.) animos eorum; in (7c 3b) rigidum cornu.

(7) a. hoc demum proelium Samnitium res ita infregit, ut omni- bus conciliis fremerent minime id quidem mirum esse, si impio bello et contra foedus suscepto. (Liv. 8, 39, 10, 1)

This defeat, after all that had gone before, so broke the spirit of the Samnites, that in all their councils they began to murmur that it was no wonder if they met with no success in an impious war, undertaken in violation of a treaty . (Trans. Foster)

b. non tamen ita infregit animos eorum, ut absisterent impe- rio. (Liv. 38, 16, 14, 4)

et he did not cow them so thoroughly that they refrained from e ercising their power . (Trans. Sage)

c. rigidum fera dextera cornu

dum tenet, infregit, truncaque a fronte revellit (Ov. met.

9.86)

Holding my tough horn in his pitiless right hand, he broke it off and tore it from my forehead, mutilating me . (Trans. Miller)

The structures in (8), which provide e amples of effringo (to break open, to break in pieces), are syntactically and semantically similar to those in (7), a P-Nom. is the originator of the change and a P-Acc.

is the theme that changes state: in (8a), illi; in (8b), illi qui maiorem habere vim credunt ea; in (8c), Philosophus primus; and the Ps- Acc.: in (8a), foris aedis; in (8c), arta claustra portarum naturae.

7 Hercules is not mentioned in this paragraph, but Ovid is referring the story of Hercules and Aqueloo.

(14)

Cadina Palachi 23 8

(8) a. Illi noctu facta manu armataque veniunt, foris aedis ef- fringunt. (Cic. II Verr. 4, 96, 10)

They, having prepared and armed a body of men, come by night; they break in the doors of the temple . (Trans.

onge)

b. qui maiorem habere uim credunt ea quae non habent artem, ut effringere quam aperire, rumpere quam solue- re, trahere quam ducere putant robustius. ( uint. inst. 2, 12, 1, 4)

Critics who think that true vigour is all the greater for its lack of art, regarding it as a special proof of strength to force what might be opened, to break what might be untied and to drag what might be led . (Trans. Butler)

c. sed eo magis acrem

inritat animi virtutem, effringere ut arta

naturae primus portarum claustra cupiret. (Lucr. 1, 69) Rather all the more they roused the ardor of his courage and make him long to be the first to burst the bolts and bars of nature’s gates . (Trans. Smith)

The e amples of the latin verbs irrumpo (to force one’s way into, to

rush suddenly into) and erumpo (to burst out, to break out) show

different structures and meanings. The P-Nominatives are: in (9a) ille,

in (9b) argumentatio, in (9c) sensus optimi. These Ps cannot be

interpreted as causing a change of state in an entity e pressed by a P-

Acc., they undergo a CHANGE OF LOCATION. In (9a), the only

e ample where we could find a P-Acc. with this verb, we can see that,

when it is in the conte t of the prefi ed irrumpo and erumpo, the P-

Acc. is not the entity that changes its state. It is clear that portas apertas

(the open doors; the entries not covered) are not entities that prevent the

passage and must be broken , but the GROUN through which the

object (ille) moves. In (8b), there is a complement that also indicates

(15)

the GROUN or point of arrival: in pectora; and in (8c) there is no P- Acc. or PPs complement.

(9) a. ille quidem potuit portas inrumpere apertas, sed nos obstitimus. (Ov. met. 15, 598)

He might have forced his way within your gates, for they stand open, but I have hindered him . (Trans. Golding) b. quam subtilis et crebra argumentatio, ut denique do-

minetur in adfectibus atque in pectora inrumpat. ( uint.

inst. 2, 5, 8, 6)

How subtle and fre uent are the thrusts of argument and in conclusion how the orator establishes his sway over the emotions of his audience, forces his way into their very hearts (of the jury) . (Trans. Butler)

c. Inrumpunt enim optimi nonnumquam sensus, quos neque inserere oportet. ( uint. inst. 10, 3, 33, 3)

For sometimes the most admirable thoughts break in upon us which cannot be inserted in what we are writing . (Trans. Butler)

The e amples in (10), in which the le eme erumpo appears, show that the Ps-Nom. are the ones that change location or move from a starting point: in (10a), flumen universum; in (10b), eloquentia; in (10c), quae (lacrimae); in (10d), audacia. The prefi ation with ex-, to say it more appropriately, marks the e it point and, therefore, the direction.

There are no Ps-Acc.: in (10a), unde indicates the starting point; in

(10b), per obstantia marks the GROUN that is being traversed; in

(10c), dolore and, in (10d), ex avaritia are the sources or the origin

where the movement begins.

(16)

Cadina Palachi 240

(10) a. sed unde universum flumen erumpat; qui illi sedis et qua- si domicilia omnium argumentorum commonstret. (Cic.

de orat. 2, 162, 8)

But to the source whence a whole flood gushed forth; to an instructor who would show him the seats and aboded, as it were, of every sort of arguments and would illustrate them briefly (Trans. J.S. Watson)

b. nondum enim tantum dicendi lumen accesserat ut etiam per obstantia erumperet. ( uint. inst. 12, 9, 5, 5)

For it had not yet attained that splendor of diction which makes it impossible to hide its light under a bushel . (Trans. Butler)

c. Quin etiam lacrimas iis natura mentis indices dedit, quae aut erumpunt dolore aut laetitia manant. ( uint. inst. 11, 3, 75, 5)

And further, nature has given them tears to serve as interpreters or our feellings, tears that will break forth for sorrow of stream for very joy . (Trans. Butler)

d. In urbe luxuries creatur, ex luxuria exsistat avaritia ne- cesse est, ex avaritia erumpat audacia, inde omnia scele- ra ac maleficia gignuntur. (Cic. S. Rosc. 75, 7)

In a city, lu ury is engendered; avarice is inevitably produced by lu ury; audacity must spring from avarice, and out of audacity arises every wickedness and every crime . (Trans. onge)

3.1. Partial conclusions and analysis

The analysis of the e amples presented allows us to observe that the

le emes derived with prefi es show a difference in meaning, and that

(17)

what happens with the Root RUMP is also different compared to what happens with the root FRANG. In the latter, the meaning of the resultant state (broken) is conserved, but in the former the meaning of the direction or movement, proper to the prefi , is favoured. RUMP (when prefi ed with in or with ex) can enter the synta with a type of P-Acc. that e presses the GROUN (the limit to cross: portas apertas); or they can take a PP-in ex. In contrast, FRANG prefi ed with ex or in, still e presses the change of state, and the P-Acc. is the Theme. We understand that the incidence of the prefi is closely linked to where the change of state takes place, rather than to the direction of the process.

The change of location, which is e pressed by the prefi with the Root RUMP, is related to the irection of the movement, and implies an argument related to the Ground (towards or from) that can be realized in the synta as a PP-in/ex a P or as a P-Acc., or not being realized. We will assume that the Roots FRING and RUMP are inserted into structures corresponding to the change of state and the change of location respectively. So, both the Roots and the arguments are interpreted according to the place they occupy in these structures.

The P in nominative of the infringo and effringo verbs is the

originator, the P in accusative is the Figure that changes state; in the

corresponding structure there is no PP P that e presses the GROUN .

The le emes prefi ed with the Root RUMP are verbs of change of

location, the P in nominative is the Figure (the element that moves),

there is no originator. The Root RUMP provides an encyclopedic

meaning linked to the (violent sudden) way of entering or leaving.

(18)

Cadina Palachi 242

vP Illi v’

v pP foris aedis p p pP

p FRING E

vP v pP

v RUMP

audacia p p pP E

p NP-Abl e avaritia Illi effringunt foris aedis

(change of state)

Audacia erumpit ex avaritia (change of location)

According to the previous analysis, that is inspired in the analysis of the English verb break in proposed by Acedo-Matellán – Mateu (2014), RUMP is inserted into structures in which the vP does not project a specifier, compared to what happens with FRING, and this is the reason why the P in nominative is not the originator but the specifier Figure of pP.

The le emes irrumpo erumpo are verbs of change of location, while

infringo effringo are verbs of change of state. Because the Root

FRING is inserted in this position, it receives interpretation of

result ; while RUMP gives phonological incarnation to a verb of

movement. The insertion of the preposition or prefi is different in each

verb. We believe that ex-/in- are inserted at the lowest p when they

e press location and, when they e press direction, they are inserted at

the highest p instead.

(19)

4. P

AST

P

ARTICIPLES

At this point, we think it is necessary to ask whether there s a possibility of having the simple verbs (frango and rumpo) in their participial forms (specifically in the past participle form

⁸

: fractus, ruptus); and what kind of elements (the P that agrees with the participle) can be in that state (ruptus or fractus). Another important uestion is what happens with the prefi ed verbs (infringo, effringo, irrumpo, erumpo) and the possibility that these verbs have to appear in the past participle form.

We think it is important to see if the passive forms of the prefi ed verbs, especially irrumpor or erumpor, do e ist in Latin and, if they do, as it seems to be the case for erumpitur, what meaning does it have.

The OLD records an entry for fractus -a -um ( broken, uneven, rugged ) as an adjective and, on the contrary, there is no entry for ruptus -a -um. However, a search in the corpus has allowed us to observe that this le eme can be found in the database not only as part of the passive periphrasis of perfect: verb sum past participle, but also agreeing with nouns (vinculis, catenis, foedus).

(11) a. namque M. Fabius, captiuus Romanus, cum per negle- gentiam custodum festo die uinculis ruptis per murum inter opera Romanorum, religata ad pinnam muri reste suspensus, manibus se demisisset. (Liv. 8, 16, 9, 2)

For Marcus Fabius, a Roman prisoner, being neglected by his guards on a day of merry-making, broke his bonds, let himself down by the wall, hand over hand, into the Roman works, by a rope which he had made fast to a battlement . (Trans. Foster)

8 It is worth to note that we are interested in the past participle because it has a passive meaning.

(20)

Cadina Palachi 244

b. ut sese inclusit ruptisque immane catenis deiecit saxum. ( erg. Aen. 8, 225)

Just as he’d shut himself in and had broken the chains he had crafted . (Trans. Ahl)

c. tum ex legatis unus abiens ‘et haec’ inquit, ‘sacrata quercus et quidquid deorum est audiant foedus a uobis ruptum’. (Liv. 3, 25, 8, 3)

Thereupon one of the envoys said, as he departed, Let both this sacred oak and whatever gods there are hear that the treaty has been broken by you . (Trans. Foster)

It is interesting to note that the noun they agree with is the entity (material or immaterial) that has become broken as a result of an action. This meaning is most clearly seen in Livy’s e ample (10c), in which a vobis is the cause or originator. So, the root RUPT enters to form Achieved States (or Results).

(12) a. equi pars in mari fractis nauibus absumpti, parti neruos succiderunt in litore Macedones. (Liv. 44, 28, 14, 2)

Some of the horses were destroyed in the sea as the ships were wrecked, some were hamstrung on shore by the Macedonians . (Trans. Schlesinger)

b. contentio fratrum trium turpis, fracta vis, contemptus fu- ror. (Cic. Att. 4, 3, 4, 8)

And the three brethren’s struggle ended in disgrace, their strength broken, and their mad pride humbled . (Trans.

Winstedt)

(21)

c. mutum forum, elinguem curiam, tacitam et fractam civitatem videbatis. (Cic. post red. 6, 9)

This body issued no declarations by its authority; you saw the forum silent the senate-house mute, the city dumb and dispirited . (Trans. onge)

In the e amples of (12) there is also an entity that has transformed to the state named by the Root FRACT. The uestion that should be asked, then, is whether the Roots FRACT and RUPT with the prefi es in- and ex- are in the database and, in that case, with which meanings. The search in PHI5

⁹

shows that the forms (of the le eme) infractus -a -um ( broken, weaken ) and effractus -a -um ( broken in pieces ) are possible and e isting. We found e amples of infractus -a -um (omnibus hastis infractis, infractos animos, oratio infracta, infrac- tus furor) and effractus -a -um (effractoque inlisit in ossa cerebro, ac ante paulo lucem effractam portam urbem ingreditur, effracta claustra portarum) which has allowed us to observe that the N that agrees with these derived forms is the one that changes state and becomes broken, or at a near point on the degradation scale.

(13) a. maestusque per omnis

it gemitus, torpent infractae ad proelia vires. ( erg. Aen.

9, 499)

Sweeps through the ranks; battle energy breaks and begins to grow sluggish . (Trans. Ahl)

b. sternitur infelix Acron et calcibus atram

tundit humum exspirans infractaque tela cruentat. ( erg.

Aen. 10, 731)

Acron goes down, unfulfilled; and his heels, in his death- throes, pound earth. His blood dyes the now broken spear that has killed him . (Trans. Ahl)

9 PHI 5 is the data base of the Packard Humanities Institute (1997).

(22)

Cadina Palachi 246

On the contrary, there are no e amples of irruptus -a -um. As for the le eme erumpo, the form eruptos

¹⁰

is found in an e ample of Lucretius, but its meaning is not that which is in a broken state but that what comes out e its :

(13) et hic Aetnaea minantur murmura flammarum rursum se colligere iras, faucibus eruptos iterum vis ut vomat ignis

ad caelumque ferat flammai fulgura rursum. (Lucr. 1, 724) Here rumbling Etna threatens to reassemble her angry flames in readiness to belch fire from her throat in another violent eruption and once more shoot to the sky flakes of flame . (Trans. Smith)

The fact that the prefi ed verbs, especially irrumpo, are not able to enter in past participle constructions, helps us to understand that, in effect, the le emes formed by prefi ing in- and ex- rumpo have been detransitivized. The Root RUPT does not appear in this structure as a result, but indicates a change of location. We have also looked in the database for the presence of these verbs in passive forms and, again, we have found e amples of the derivatives of FRING- (15a and b) and two of the verbs with the root RUMP and the prefi ex- (16a and b), but none of irrumpo.

10 Acedo-Matellan (2010: 118), when discussing the Complement irect Motion Construction, writes: I note, last, that a possible countere ample to the deponent- restriction on past participle availability in intransitive verbs is provided, funnily enough, by a non-motional base: RUMP break . This root yields C MCs when accompanied with a directional e pression, and it contributes, as happens in English predicates involving break particle, a manner component can be paraphrased as

in an abrupt, harsh or violent way .

(23)

(14) a. quae si in extremo breviora sunt, infringitur ille quasi verborum ambitus; sic enim has orationis conversiones Graeci nominant. (Cicero de orat. 3, 186, 9)

And if they are shorter at the end, this makes a break in the periodic structure of the words – for ‘period’ is the Greek name for these turning-points of speech . (Trans.

Rackham)

b. sed surdae ad omnia aures concitatae multitudinis erant nec minore intus ui quam foris portae effringebantur, effractisque omnibus toto Hexapylo agmen receptum est.

(Liv. 24, 32, 6, 3)

But the ears of the e cited crowd were deaf to all that;

and the gates were being forced with no less violence from within than from without, ad when all had been forced, the column was admitted through the whole breadth of the He apylon . (Trans. Moore)

(15) a. post incita cum vis exagitata foras erumpitur et simul altam

diffindens terram magnum concinnat hiatum. (Lucr. 6, 583)

Then its impetuous force, lashed to fury, bursts out and, in so doing, cleaves the earth to its depths and opens a yawning chasm . (Trans. M.F. Smith)

b. ceterum, ubi primum sub ictu teli fuerunt, duabus simul portis erumpitur; et ad clamorem erumpentium ingens strepitus e muris portus ululantium mulierum. (Liv. 43, 10, 5, 3)

However, as soon as they were within range of missiles, a sally was made from two gates at once; and together with the battle-cry of the sallying parties, there arose from the walls a great uproar of women howling . (Trans.

Schlesinger)

(24)

Cadina Palachi 248

5. C

ONCLUSIONS

The rumpo and frango le emes, formed from the Roots RUMP and FRANG can be classified as belonging to the class of CHANGE OF STATE. The verbs, resulting from the syntactic insertion of the Roots in the corresponding structures, are transitive and their direct complement is the entity whose state is modified.

The Roots themselves, in the case of frango the Root FRING, when inserted with the prefi es in- and ex- behave differently from each other. RUMP only retains a kind of MANNER flavor. The resulting verb is one of movement and has the meaning CHANGE OF IRECTION, the Figure is the syntactic subject and the semantic theme, since it is the object that moves. The Root provides a meaning linked to the manner of the movement (with some violence or sudden).

On the other hand, when FRING is inserted with the same prefi es, its meaning is still CHANGE OF STATE, the prefi es have LOCATION values. The structure is transitive, so that the P-Nom (syntactic subject) is the originator of the Change, the direct complement ( P- Acc) is the Theme that changes state.

Our work tries to e plain this behavior from a theory that understands that the Roots are inserted into different syntactic structures, and it is the structures that provide the semantic features and the types of arguments re uired. The Roots provide the meanings, which can be forced to change due to functional phrases.

We would like to continue this research in order to understand what differentiates, e actly, the Roots in order to make them able to change their meanings when prefi ed.

Universidad Nacional del Litoral

FHUC – IHuCSo Santa Fe (Argentina)

[email protected]

(25)

B

IBILIOGRAPH

Acedo-Matellan, .

2010 Argument structure and the syntax-morphology interface.

A case study in latin and other languages, Barcelona, Universidad Autónoma de Barcelona, octoral issertation.

Acedo-Matellan, . – Mateu, J.

2014 From syntax to roots. A syntactic approach to root interpretation, in Ale iadou, A. – Borer, H. – Sch fer, F. (edd.), The syntax of roots and the roots of syntax, O ford, O ford University Press, pp. 14-32.

Ale iadou, A. – Borer, H. – Sch ffer, F. (edd.)

2014 The Roots of Syntax and the Syntax of Roots, O ford, O ford University Press.

Alvarez Huerta, O.

2009 Acusativo, in Ba os Ba os, J.M. (ed.), Sintaxis del latín clásico, Madrid, Liceus, pp. 131-154.

Arad, M.

2005 Roots and Patterns. Hebrew morpho-syntax, Stanford, Springer.

Borer, H.

2005 The normal course of events. Structuring sense, O ford, O ford University Press.

2013 Structuring Sense. Taking Form (Vol. III), O ford, O ford

University Press.

(26)

Cadina Palachi 25 0

Embick, .

2005 The Morpheme, Boston, Walter e Gruyter.

2009 Talk at the Roots Workshop. Roots, States and Stative Passives, Stuttgart.

Embick, . – Noyer, R.

2007 Distributed Morphology and the Syntax-Morphology interface, in Ramchand, G. – Reiss, C. (edd.), The Oxford Handbook of Linguistics Interfaces, O ford, O ford University Press, pp. 287-324.

2014 Roots and Phases, in Ale iadou, A. – Borer, A. – Sch fer, F. (edd.), The syntax of roots and the roots of syntax, O ford, O ford University Press, pp. 192-207.

Glare, P.

1968 Oxford Latin Dictionary, O ford, Clarendon Press.

Hale, .L. – eyser, S.J.

1987 A View from the Middle, Cambridge (MA), Center for Cognitive Science, MIT, Le icon Project.

Halle, M. – Marantz, A.

1993 Distributed Morphology and the Pieces of Inflection, in Hale, . – eyser, S.J. (edd.), The view from Building 20. Essays in Linguistics in Honor of Sylvian Bromberger, Cambridge (MA), The MIT Press, pp. 11-176.

Harley, H.

2014 On the identity of roots, in Theoretical Linguistics , pp.

225-276.

(27)

Levin, B.

1993 English Verb Classes and Alternations. A preliminary investigation, Chicago, The University of Chicago Press.

Marantz, A.

1997 No Escape from Syntax: Don’t try morphological analysis in the privacy of your own lexicon, in imitriadis, A. (ed.), Proceedings of the 1998 Penn Linguistics Colloquium, Philadelphia, Penn Working Papers in Linguistics.